Ray Proudfoot

BSOD - BAD_POOL_HEADER - glitch or something else?

Recommended Posts

Twenty minutes into a flight from Birmingham (EGBB) to Ponta Delgada (LPPD) in Concorde my flight sim machine suffered a BSOD with BAD_POOL_HEADER shown as the cause. Sim is P3D v3.4. Computer specs in sig,

It appears to be memory related. On 8 June I replaced my existing RAM with two new Crucial sticks mainly as preparation for P3D v4. When I first booted it up after replacing the memory I noticed a LED code on the mobo - A0 which equates to IDE initialization is started but despite leaving the BIOS alone this doesn't seem to have been an issue. The PC has been very stable since I bought the memory and I've flown perhaps 30-40 flights all without a problem.

I rebooted the PC, performed a Windows Memory Check and it found nothing. Loaded the flight from an AutoSave and it was fine for the remainder of the flight.

Is this just a one-off glitch or something more worrying? .

Might that LED code be telling me to do something? The PC was professionally built and overclocked so I don't really want to go into the BIOS.

PS: Note to Mods. I originally posted this in the Windows7 forum but had no replies. Can you delete that post please?

Share this post


Link to post
Share on other sites
Help AVSIM continue to serve you!
Please donate today!

I've created a bootable USB with MemTest86 and started all tests.

Immediately it showed 32 errors and after just a couple of minutes it now stands at 2316.

This is Corsair Vengeance memory - 2x 8Gb sticks as recommended by Scan UK as my existing memory was no longer available. The system has been professionally overclocked by Scan but of course when running MemTest that is ignored.

Have I failed to do something in the BIOS? It seems ridiculous that so many rrrors could be found in new memory.

Share this post


Link to post
Share on other sites

Running MemTest on sticks separately. First stick showed multiple errors almost immediately. Put other stick in same slot and now on pass 2 of 4 and no errors. Looks like bad module which I'm sure Scan will swap when I speak to them tomorrow.

Share this post


Link to post
Share on other sites

As a Scan customer of long standing, Ray, I can vouch for their excellent Help Desk service,so hopefully, this will be resolved. I had that odd stick of RAM go down, with OCZ RAM( no longer in the memory business), too.

Share this post


Link to post
Share on other sites

Hi Ray,

I can only sympathise with your issue, I am on my second pair of CorsairVengeance Pro DDR3 2x8Gb - 16Gb 2400Mhz modules. Only a few months after purchase I was getting many BSOD's so checked the memory with Memtest86+ (CD-boot) one module at a time, as it turned out both modules had many errors one had a (84) and the other had many thousands of errors (3891). I noticed the heat spreaders were loose and the thermal pads glued to the memory chips had become detached. When I asked for a reason my expensive modules had started to come apart and show errors they just said Corsair modules can be like that, maybe they were made on a Friday afternoon of a long weekend.

Anyway long story short, I had to RMA them through my system builder for matched perfect replacements, the modules were hard to come by as this was around the time DDR4 modules were becoming popular and stock was not readily available for the exact replacement kit (I think I waited 2 months here in Australia).

The replacement Kit was tested the moment it arrived Memtest86+ 10 passes 0 errors and still going strong 3 years later.

ac9292ede1a0dfbcfaf76ae880bfc59f.jpg
 

I am Running an ASUS Z87 Expert motherboard which has the ASUS Motherboard LED numerical Display, as far as I can tell A0 would be correct once booted into the OS, I would expect Gigabyte to be the same/similar. You would then see the display change when EIST sleep states apply for Sleep or Hibernate OS condition.
If your new Memory modules were being miss-identified or failing to be recognised/initialised the LED codes would indicate this with a specific code, ASUS codes are 50-53, 54, 55, at which time you can press the DRAM OK button. (Not sure what Gigabyte have) 
Indeed A0 indicator is identifying that (IDE Initialization is started) this would seem to be the correct display. A0 = A0K IDE state has handed over successfully to OS for booting.

(Edit) Note: when you talk to your supplier tomorrow you should ask for a replacement kit as they will have been matched specifically, don't replace only the error'd module alone, get a matched Kit with identical specification. 

Because the new Memory has a different spec to your original (Crucial)modules, you will need to set this in bios, you may also need to reset the XMP profile, of course this will also affect your current overclock which may or may not be improved with the new modules. When you reset your RAM, you will also need to reset your Overclock.

Best of Luck I hope your replacements are as good/reliable as mine have been.

Cheers Jethro

  • Upvote 1

Share this post


Link to post
Share on other sites

Hi Jethro / Rick,

Thank you both for your replies. Just back from watching Dunkirk (excellent!) and the test of module 1 has completed. Zero errors. Great.

I'm now testing it in the other slot to ensure the error lies in the memory and not the mobo. I shall know in a few hours.

Good point about a replacement kit. I'll ask them tomorrow about that. Regarding setting the BIOS I didn't need to change anything when I swapped my old memory for these new ones. Voltage was okay and overclocking was same as before.

I shall also ask them if they test memory before dispatching as I don't want the same problem again.

More tomorrow.

Share this post


Link to post
Share on other sites
1 hour ago, Ray Proudfoot said:

I'm now testing it in the other slot

The smart way to determine it is not the memory controller on the MoBo, Ray. Fingers crossed it's not the MoBo.

Share this post


Link to post
Share on other sites
2 hours ago, vc10man said:

The smart way to determine it is not the memory controller on the MoBo, Ray. Fingers crossed it's not the MoBo.

It's not Rick. Testing the good module in both slots returned zero errors. :biggrin:

I've now put the bad module in (good one removed) and am letting the test complete to get a total count of errors. I'll use that tomorrow to ask Scan to send me a matched tested pair and I'll return the faulty pair on receipt.

Share this post


Link to post
Share on other sites

I tested the bad module in both slots yesterday. The process was unable to finish because of errors. Total error counts were 21,088 and 19,346. :ohmy:

Spoke to Scan this morning and they advised to return both modules and on receipt they will send out by return a replacement pair. Should have it by end of week or maybe early next.

At least I can use my old 2 x 4Gb pair in the meantime.

Share this post


Link to post
Share on other sites
12 hours ago, Ray Proudfoot said:

It's not Rick. Testing the good module in both slots returned zero errors

You may have misunderstood my theory Ray. What I meant to say was, if both memory sticks returned no errors in one RAM slot, and then returned errors in the other RAM slot, surely that would point to a faulty motherboard memory controller.

Share this post


Link to post
Share on other sites

But both didn't Rick. One module was clean and the other had thousands. And I tested both in both memory slots just to confirm the problem lay with the memory, not the mobo.

Share this post


Link to post
Share on other sites

Ray, hopefully, Scan have it in stock and will replace both modules asap.

Share this post


Link to post
Share on other sites

Rick,

I forgot to ask them if stock was held but it's a pretty standard memory so hopefully it will be. In any case I've put my old 8Gb of memory back in, MemTested it and 0 errors so I'm fine for a week or more.

They wanted both modules back which makes sense given they come in pairs.

Share this post


Link to post
Share on other sites
1 hour ago, Ray Proudfoot said:

come in pairs.

I had to do the same with the now defunct MicroDirect when my OCZ Reaper RAM modules went south. Luckily, they had stock when I visited their sales place opp McVities.

Share this post


Link to post
Share on other sites

I wondered what happened to them. I bought a Nvidia 8800GTX 768Mb card from them yonks ago and still have it running in a support PC. Does a great job.

Share this post


Link to post
Share on other sites

That genre of 8800GTX cards were one of Nvidia's better cards for the then time. That's why it still runs well.

Share this post


Link to post
Share on other sites

I swapped a top-end faulty Radeon for it. radeon had nicer PQ but was rubbish with clouds. 8800GTX took them in its stride.

Share this post


Link to post
Share on other sites

I too got rid of a Radeon that was the recommended card in all technical mags a the time, then, which then died a death on me, at around the time these Forums were full of FS2004 topics that suggested Nvidia cards were better.

Share this post


Link to post
Share on other sites

Scan have tested the memory and confirmed it's faulty. Replacement modules sent out today. Should have them Monday.

Good service from a good company. :smile:

Share this post


Link to post
Share on other sites

Excellent news Ray, things are looking up!   21000 errors on the faulty module, you must have seen an occasional weird OS behaviour. Hope your system becomes rock solid and stable with the new RAM.

Cheers Jethro

Share this post


Link to post
Share on other sites

Strange, I replied to your good news, Ray and my post got deleted. Wonder why?

Share this post


Link to post
Share on other sites
16 hours ago, Jethro said:

Excellent news Ray, things are looking up!   21000 errors on the faulty module, you must have seen an occasional weird OS behaviour. Hope your system becomes rock solid and stable with the new RAM.

Cheers Jethro

That's the odd thing. I didn't. I can only assume the first module was taking most of the load. In 6 weeks I had just the one BSOD. Abut it was significant enough to make test the modules and the rest is history.

Rick it would be odd if a post was removed without you being notified.

Share this post


Link to post
Share on other sites
7 minutes ago, Ray Proudfoot said:

Rick it would be odd if a post was removed without you being notified

Does not really matter, Ray. I was merely following up the news you received from Scan complimenting them on their excellent service. If somebody saw fit to delete it, so be it. I will not lose sleep. C'est la vie..........

Share this post


Link to post
Share on other sites

The replacement modules arrived today and after a 3 hour MemTest I'm pleased to report zero errors.

Should anyone else get this BSOD message I hope they find this topic helpful.

Share this post


Link to post
Share on other sites

Brilliant news, Ray, and thanks for the update.

Q.E.D.??(if you are like me, an old timer)

  • Upvote 1

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now